Prodigy Annotation

Prodigy is an annotation tool built by the creators of spaCy that uses active learning to intelligently select the most informative examples for human labeling, minimizing the total annotation effort needed to train high-quality NLP and computer vision models. Its scriptable annotation workflows support named entity recognition, text classification, image segmentation, audio labeling, and custom annotation tasks. Prodigy runs locally for complete data privacy and integrates seamlessly with spaCy and PyTorch training pipelines for efficient model development cycles.